Instructions to use SargeDev/Solar-Decision-JEV-27B-V3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SargeDev/Solar-Decision-JEV-27B-V3 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="SargeDev/Solar-Decision-JEV-27B-V3")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SargeDev/Solar-Decision-JEV-27B-V3")
model = AutoModelForCausalLM.from_pretrained("SargeDev/Solar-Decision-JEV-27B-V3", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use SargeDev/Solar-Decision-JEV-27B-V3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SargeDev/Solar-Decision-JEV-27B-V3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SargeDev/Solar-Decision-JEV-27B-V3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SargeDev/Solar-Decision-JEV-27B-V3

SGLang

How to use SargeDev/Solar-Decision-JEV-27B-V3 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "SargeDev/Solar-Decision-JEV-27B-V3" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SargeDev/Solar-Decision-JEV-27B-V3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "SargeDev/Solar-Decision-JEV-27B-V3" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SargeDev/Solar-Decision-JEV-27B-V3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SargeDev/Solar-Decision-JEV-27B-V3 with Docker Model Runner:
```
docker model run hf.co/SargeDev/Solar-Decision-JEV-27B-V3
```

Solar-Decision-JEV-27B-V3

The V3 solar run: combined corpus + r32 geometry + lm_head readout. huihui Qwen3.8-27B-abliterated + our solar LoRA (r32/α64) trained on the union of jev-distill-corpus-v3 + solar-decisions-corpus-v4 (764,626 rows, 90 domains) with lm_head added to the target modules, merged at step 2184.

This is the third generation of SargeDev/Solar-Decision-JEV-27B. Two changes vs V2 (r32+step-1894-solar):

Training data: V2 trained the solar half on solar-decisions-v4 alone. V3 trains on the union — the r32-style verbalizer/decision-shape rows are in the mix, attacking the readout gap at the data level.
Target modules: V2 targeted attention+MLP only. V3 adds lm_head, so the decision head's output vocabulary geometry is trained alongside.

The decision readout is trained INTO the model — a typed decision is a single prefill with max_tokens=1, read back as log-probabilities over the option tokens. No separate head to run. (vLLM only — llama.cpp does not return the per-token logprobs this path needs; use the GGUF for chat, not for decisions.)

Gate results, honestly (27,695 rows, same harness/head/calibration as every number below)

metric	V2 (r32+1894)	V3 (step 2184)	r32-alone ref	AutoTrust bar
choice top-1	0.571	0.833	0.869	0.904
score top-1	0.567	0.794	—	0.891
noul AUROC	0.962	0.988	0.992	0.996
noul top-1	0.879	0.939	—	—
choice ECE	0.021	0.167	—	—

(below bar) on choice/score top-1: V3 closes ~⅔ of the gap V2 left to r32-alone, but does not reach the bar. The sharp rise in choice ECE (0.021 → 0.167) means the model is much more confident on hard rows — better ranking, worse calibration. If you need calibrated probabilities, prefer r32-alone (SargeDev/JEV-27B-Uncensored); if you need top-1 accuracy from a self-contained model, V3 is our best.

What's in the repo

bf16 merged weights (19 shards, 52 GB) — decisions (vLLM) + everyday chat
decision_head.json + calibration.json + serve_decide.py — the typed-decision machinery (credit AutoTrust AI)
MTP/nextn drafter tensors preserved (speculative decoding works in llama.cpp)

Quantizations

NVFP4-FP8: SargeDev/Solar-Decision-JEV-27B-V3-NVFP4-FP8 (29 GB)
GGUF ladder (11 quants, f16 + Q2_K … Q8_0): SargeDev/Solar-Decision-JEV-27B-V3-GGUF

llama-bench (GB10, CUDA build, `-ngl 99`, pp512 / tg128)

quant	size	pp512	tg128
Q4_K_M	15.65 GiB	865.94	12.80
Q8_0	27.04 GiB	818.03	8.38

Lineage & credits

huihui-ai — abliterated Qwen3.8-27B backbone
AutoTrust AI — r32 decision LoRA, 24-slot decision head, calibration, serve_decide.py
SargeDev — jev-distill-corpus-v3, solar-decisions-corpus-v4 (the training data)
Qwen — base model
TypeSafe AI — Jev 1.13 teacher (referenced, not redistributed)

Downloads last month: 210

Safetensors

Model size

27B params

Tensor type

BF16

Model tree for SargeDev/Solar-Decision-JEV-27B-V3

Base model

Qwen/Qwen3.8-27B

Finetuned

huihui-ai/Huihui-Qwen3.8-27B-abliterated

Finetuned

(15)

this model

SargeDev
/

Solar-Decision-JEV-27B-V3

Solar-Decision-JEV-27B-V3

Gate results, honestly (27,695 rows, same harness/head/calibration as every number below)

What's in the repo

Quantizations

llama-bench (GB10, CUDA build, `-ngl 99`, pp512 / tg128)

Lineage & credits

Model tree for SargeDev/Solar-Decision-JEV-27B-V3

Datasets used to train SargeDev/Solar-Decision-JEV-27B-V3

Solar-Decision-JEV-27B-V3

Gate results, honestly (27,695 rows, same harness/head/calibration as every number below)

What's in the repo

Quantizations

llama-bench (GB10, CUDA build, -ngl 99, pp512 / tg128)

Lineage & credits

Model tree for SargeDev/Solar-Decision-JEV-27B-V3

Datasets used to train SargeDev/Solar-Decision-JEV-27B-V3

llama-bench (GB10, CUDA build, `-ngl 99`, pp512 / tg128)